The Invulnerable Pleasures of Epicurean Friendship

O'CONNOR, DAVID K., The Invulnerable Pleasures of Epicurean Friendship , Greek, Roman and Byzantine Studies, 30:2 (1989) p.165

The Invulnerable Pleasures of Epicurean Friendship

David K. O'Connor

E PICURUS DEVELOPED a distinctive account of the value of friendship, an account he surely intended to be consistent with his more general hedonic account of value and hap

piness. Many critics both ancient and modern believe that Epicurus failed, with diagnoses of the failure falling into two camps. One camp claims that Epicurus' attachment to the view that pleasure is the sole constituent of hapriness permitted him only a mercenary and niggardly conception 0 friendship. They charge that the sort of friendship he held up as an ideal for his disciples falls short of the highest friendship, especially because it entails too little sacrifice of one's own interests for one's friends'. Critics from the other camp deny that Epicurus' ideal is insufficient, and in fact often praise him for his generosity on this point. But they find this generosity inconsistent with his hedonism. According to a recent critic, "In his account of friendship, Epicurus seems quite willing to concede the common belief that friendship requires valuing others for their own sake, even though this creates difficulties for his hedonism."l

So the critics view Epicurus either as consistent philosopher and bad friend, or as inconsistent philosopher and good friend. Either way, criticism focuses on what Epicurus' hedonism excludes from friendship, while taking for granted what a good friendship is. I think it is fairer to Epicurus, and more interesting philosophically, if we focus instead on what he believed friendship is and could be. He does not merely give us novel hedonic reasons for living up to an already given ideal of friendship. Instead, Epicurus offers us an alternative ideal, and invites us to measure our friendships against his

1 P. MITSIS, "Epicurus on Friendship and Altruism," aSAP 5 (1987: hereafter 'Mitsis'J) 126-53, at 131 n.ll. A slightly revised version appears as chapter 3 of his Epicurus' Ethical Theory: The Pleasures of Invulnerability (Ithaca 1988) 98-128. My references are to the original article.

165


166 EPICUREAN FRIENDSHIP

new standard. He has not claimed simply to have found a shorter path than his opponents to the top of the same old mountain, but to have found a new and more lofty peak. He challenges us to reorganize our experience of friendship by putting the exquisite pleasure of peaceful tranquillity at its heart. We shall first try to reconstruct in some detail what such friendship would be (section I).

Once we have seen more clearly what the ideal Epicurean friendship asks of us and offers to us, we can better confront the criticisms leveled against it. I then (section II) consider a question that arises immediately from Epicurus' own concerns: does living up to this ideal of friendship undermine our self-sufficiency and control over our own happiness? Recently, the notion has become popular that if love and friendship are indispensable elements of a happy life, happiness must be more vulnerable to chance than many Greek moral philosophers were willing to concede. 2 I argue that this is not true of Epicurus' ideal friendships. We neither depend on spontaneous, fortuitous affections for their formation nor become vulnerable to unlucky pains at their dissolution. The link between self-sufficiency and tranquillity is preserved when we live together as Epicureans should.

A second question, less central within Epicurus' own perspective but important to many critics both ancient and modern, asks whether the Epicurean ideal of friendship is too undemanding in what it asks friends to give up for each other. The ancients raised this question by asking whether a consistent Epicurean would be willing to die for a friend. This particular sacrifice is used here as a test of the nobility of a moral ideal: an ideal that can justify giving up so great a good as one's own life must promise even greater and nobler goods to those who live up to it. Later Epicureans apparently tried to show that their ideal would indeed justify this ultimate act of generosity to a friend. 3 But I shall suggest (section III) that an Epicurean's reasons for being willing to die for a friend undermine the very assumptions that make such a death a test of nobility. The

2 E.g. B. Williams, -Moral Luck," in Moral Luck: Philosophical Papers 1973-1980 (Cambridge 1981) and Ethics and the Limits of Philosophy (Cambridge [Mass.] 1985); M. Nussbaum, The Fragility of Goodness (Cambridge 1986).

3 See D.L. 10.121. Perhaps the mention of taking risks for friendship in VS 28 is a hint that Epicurus himself would have been interested in the question.


DAVID K. O'CONNOR 167

particular wayan Epicurean will devote him- or herself to a friend is a measure of the distance between Ericurean ideals and the ideals implicit in the test, and not at all a proo of their similarity.

Modern critics are less concerned with the nobility of Epicurus' ideal friendship than with its perceived egoism. In section IV, I shall question Phillip Mitsis' recent claim to have found a commitment to altruism (i.e., to valuing one's friends for their own sake rather than for the pleasure they provide to oneself) within Epicurus' account of friendship. So far as I can see, Epicurus never so much as entertained the possibility that altruism was necessary for the best friendship, let alone that the value of altruism was independent of his hedonic criterion. His ideal of friendship is hedonist through and through, which is to say that he values friends because they contribute so much to living a peaceful, tranquil life. This ideal excludes the sorts of selfish behavior that provoke the charge of egoism, but the reasons for this exclusion have nothing to do with valuing altruism as such.

Before we begin, we should note what sort of inquiry is possible here. My attempt to explicate an Epicurean ideal of friendship and to explore its resources for responding to criticism suffers under the same debility as its competitors: the paucity of our sources. Discussion of friendship in the surviving writings of Epicurus is slight: two of the forty Principal Doctrines, a handful of Vatican Sayings, along with some incidental discussions in the Letter to Menoeceus. There are also some reports and anecdotes of various credibility about Epicurus and his friendships that may be relevant, and Cicero preserves some evidence useful primarily for showing what Epicurus' position (as opposed to that of later Epicureans) was not. But compared with our sources for the basic Epicurean doctrine on pleasure, for example, those on friendship are meager. We are in no position to reconstruct anything like a complete theory of friendship, and must in fact be guided more by Epicurus' general views of pleasure and happiness than by any very definite constraints that his sayings on friendship would impose. Much that I say here will therefore be frankly speculative, at best an appreciation of the Master's spirit rather than an interpretation of his letter. I believe that an Epicurean ideal of friendship is nevertheless worth considering, in part simply for its value in making us reconsider what we might otherwise take for granted about friendship itself.



I. The Epicurean Ideal

Epicureans were notorious for the way they embodied their views of friendship in a distinctive style of communal living. 4 This concrete and practical way of life is a good place to start thinking about how an Epicurean friendship contributes to a happy, successfullife-just as reflection on political life at Athens or philosophical life with Socrates or within the Academy helps us to understand what Plato and Aristotle say about love and friendship. Unlike the political friendships discussed in the Nichomachean Ethics, Epicurean social life clearly did not focus on shared public action within a polis; nor does it seem to have focused on shared theoretical investigations of the sort portrayed in Plato's dialogues. Epicureans valued less the pleasures of a good colloquium than of the reception afterwards. The heart of Epicurean friendship seems to have been fellowship rather than political action or philosophical conversation.

We will feel more vividly the distinctive appeal of this Epicurean fellowship if we consider a few important aspects of their communallife. No doubt one thing Epicurean friends did together was study Epicurus' philosophy. But a more distinctive manifestation of their fellowship seems to have been a hero cult in honor of the founders of the school. 5 The central event of these regular gatherings was a common meal. Breaking bread together, rather than engaging in political action or arguing with one another, was the focus of Epicurean fellowship. These banquets were devoted to remembering the dead-both deceased members of the local community and famous Epicurean heroes, including Epicurus himself. This practice of remembrance involved the public telling of tales of exemplary biography, a sort of Epicurean Lives of the Saints. For example, the story of Epicurus' tranquillity in his hnal hours in the face .of great bodily pain might be read to the assembled commumty.

Banqueting, remembering the dead, meditating on the teachings of the Master: if we want to understand the great value Epicurus

4 There are enough sneers from outsiders to justify calling Epicurean friendships 'notorious' .

5 Here I am heavily indebted to Diskin Clay, -The Cults of Epicurus," Cron Ercol16 (1986) 11-28, which gathers and discusses the evidence for the existence of this cult. Note especially Epicurus' endowment of these memorial banquets in his will (D.L. 10.18).


DA YID K. O'CONNOR 169

placed on friendship, we should perhaps start by trying to appreciate how such quasi-religious fellowship could be so important to a tranquil and pleasant life.6 How can this style of community life be justified on hedonic grounds?

We might be tempted to think that on Epicurus' view humans have an intrinsic desire for friendship, such that friendship would provide a special, characteristic pleasure of its own, a pleasure so important and pleasing that a life lived without it would lack an essential element of pleasant tranquillity. On this view, even the wisest Epicureans, if they lived alone, would feel the pang of an unsatisfied desire for the company of others, much like what we call loneliness. They would need friends to fill what would otherwise remain a gap in their complete satisfaction. This approach would assimilate the satisfaction of the need for friends to the satisfaction of what Epicurus called the necessary desires, e.g. for food and drink.l The friendless Epicurean would be almost literally starved for affection. But though this view seems natural enough to us, there is no evidence that Epicurus believed in any such direct desire for friendship.8 Its hedonic value must be explained less directly.

An alternative view would avoid placing the importance of friendship in fulfilling so particular a need. Rather than valuing friendship in itself as the uniquely satisfying object of an inescapable desire, we might value it for its broader effects on the tranquillity of our lives. To some extent, friendship provides the same sort of bodily and mental security from thieves and cutthroats as justice does; but

6 I do not mean to claim that the only or even the most important context of Epicurean friendship was the formal celebration of the hero cult; but these central features of the cult do illustrate in a striking way why Epicureans found friendship so valuable.

7 This interpretation would also leave open the same possibilities for variation and substitution: one friend could be substituted for another as one type of food and drink can be as satisfying as another. Thus on this interpretation the Epicurean view of the need for and value of friends would be quite different from contemporary conceptions of friendship that emphasize loving individuals and deny the replaceability of a lost friend by a new one.

8 There is in fact considerable evidence that Epicurus explicitly denied any such natural impulse to community: see P. A. Vander Waerdt, '"Hermarchus and the Epicurean Genealogy of Morals, " TAPA 118 (1988) 93 n.26. The view that Epicurus believed in a direct desire for friendship might be thought to receive support from VS 23, 1tuoa <J>tA.\a St' eau'tftv aipE'tTt [Usener: apE'tTt MSS.]. apxTtV s' tlA.ll<J>EV a1tO 'til~ cOcp£A.t\a~, a saying I discuss below (IV).



clearly we should expect rather more from friendship than this. Closer to the heart of Epicurus' conception of happiness are the cozy pleasures of the temperate yet festive common banquets. 9 Let contemplative Platonic elves and practical Aristotelian dwarves pretend to 'higher' pleasures; the Epicurean will not be embarrassed by his hobbit-like appreciation of ample food, good drink, and a warm fire. But the gemutlichkeit and physical well-being enjoyed in wholesome feasting does not yet guarantee our complete satisfaction in life. For even after we remove any anxieties caused by the predations of others, even when we achieve a healthy, pleasing bodily state, we are still prey to mental fears that destroy our happiness. "'It is no profit to obtain security as far as concerns human beings," says Epicurus (KD 13), "'while suspiciously eyeing the things above and the things beneath the earth, and generally the things in the unlimited universe." We might add to these cosmic anxieties those discontents based on false judgments about the sources and limits of pleasure and pain (KD 11). Can Epicurus have expected those who live up to his ideal of fellowship to find aid against these mental disturbances much as against the physical?

There is every reason to think he did. Some of this aid probably came from the communal study of philosophy.10 Without some basic understanding of the nature of the divine and the sources and limits of pleasure and pain, no one could be safe from vain fears and discontents. The main surviving account of Epicurus' moral teaching, the Letter to Menoeceus, implies that some philosophical training was an aspect of communallife. 11 But we must not overrate the

9 Cf. Arist. Pol. 1280b36-41: religious festivals are for the sake of mere living together (<T\)~i1v) rather than noble actions (xaAaI. 1tpa~El.~).

10 Consider VS 74 and the mention of "'discussions" ('to'i~ 1tEPI. <puaECJ)~ ~1.aA.o'Y1.aJlOl.9 at VS 10.

lIThe letter concludes with an exhortation to study its contents .. night and day by yourself and with someone like you," a clear indication of group study. A. A. Long and D. N. Sedley, The Hellenistic Philosophers I (Cambridge 1987) 144, have challenged this reading, rejecting Usener's 1tpO~ ('tE), and find no reference in the passage to studying with another person. Somewhat more revealing is Epicurus' earlier warning to Menoeceus (Ep. ad Men. 132) about the scandalous accounts of Epicurean hedonism spread by ignorant or confused detractors: "'Neither non-stop drinking and partying, nor indulgence in boys and women, nor in the fish and other things lavish tables provide, makes for the pleasant life, but rather sober thinking (A.o'Yta .... o~), and searching out the grounds for all choice and avoidance, and driving out those opinions from which the greatest disturbance gets hold of souls." Epicurus seems here to be responding to the popular



importance of such study; Epicurus is quite explicit about the merely instrumental value of philosophy.12 The theoretical remains firmly subordinated to the practical. I suggest that groups in which Epicureans studied philosophy together were typically not research seminars, pushing forward the limits of knowledge; they were probably more like refresher courses to shore up any erosion in the central Epicurean commitments of the particirants.

More important than communal philosophica study, I believe, was the concrete example of the other members of the Epicurean community, illustrated in a striking way by the school's hero cult to Epicurus and its other founders. The cult was doubtless intended solely for the edification and support of the living members of the community. Cicero (Fin. 2.103) and Plutarch (MOT. 1128A-30B) held this cult inconsistent with the fundamental Epicurean dicta "Death is nothing to us" and "Live unknown," but these are criticisms that do little more than vent polemical spleen. Eficurus did not endow these cultic celebrations to guarantee his sou a vicarious enjoyment of the pleasures of the table after he was gone, nor to attain immortal fame. The cult seems to have aimed at two benefits for its celebrants: the enjoyable memory of the goodness of the departed heroes of Epicureanism, and the inspiration of their example. 13 The importance in Epicurean moral education of imitating such exemplary precursors can hardly be stressed enough. Even the Master himself found comfort and inspiration in the calm and fearless example of the many close associates who died before him.14 Those

fantasy that Epicurean communities, with their motley collection of citizen and alien, male and female, slave and free, had extravagant orgies. He replaces this false picture of Epicurean gatherings (perhaps of the cultic celebrations themselves) with a true one in which we can perhaps see the outlines of the 'sober thinking' of communal philosophizing.

12 See KD 11 and Ep. ad Men. 132 (the priority of <Pp6vllcr~ to theoretical inquiry). Long and Sedley (supra n.11: 56) claim, on the basis of VS 27, that philosophy has more than instrumental value for Epicurus; but I think this saying simply indicates that the fruit of philosophy (namely, pleasant tranquillity) grows along with philosophic understanding. The distinction between the philosophizing and its fruit should be preserved.

13 See e.g. VS 66. I am indebted to Diskin Clay's discussion, "Individual and Community in the First Generation of the Epicurean School," in I:'\)~ft'tTlcrt~: Studi sull' epicureismo greco e romano offerti a Marcello Gigante I (Naples 1983) 255-79.

14 See Clay (supra n.13); also VS 61 on the encouragement of friends.



who had traveled the Epicurean road loyally and successfully were a vivid reminder of the promise of this way of life, the surest proof of its attainability, and their exemplary biographies the surest guide for those still on the journey.

I believe that this kind of support through remembrance and example was the chief consideration behind both Epicurus' written praise of friendship and his practical arrangements for the community that looked to him for leadership. There is no great difference on this point between living and dead heroes:15 either can be the source of comfort when we doubt the value or attainability of an ideal we have been seeking. The solidarity of the community reinforces the commitment of every individual within it. To outsiders, this solidarity might sometimes look like narrow-mindedness: when Athenaeus (5.182A=fr.56 Usener) says that Epicurus'Symposium consists of a crowd of flatterers praising one another, he means to criticize more than a particular book. This negative judgment has found its way into the often repeated contrast between the stagnant orthodoxy of the Epicurean school and the dynamic development of Stoicism. 16 But from the inside, Epicurean friendships did not have this sinister air.17

We should not, then, think of such friendships as generating some special, characteristic pleasure indispensable for hedonistic happiness. What we miss if we do not give fellowship its rightful place in our lives, an Epicurean would argue, is not a distinctive and wonderfully potent kind of pleasure. Instead, we will risk undermining that secure tranquillity which is the bedrock of all pleasurable living. Supportive fellowship is the coat of varnish protecting the contented life we paint for ourselves, not some brilliant color we need to complete the picture. The protection this supportive fellowship provides is mental as well as bodily. No doubt our friends defend us from aggressors and help us in times of distress. But they

15 As Clay (supra n.13) points out, the hero cult existed before Epicurus' death, as is shown by the provisions of his will (D.L. 10.18; cf. VS 32).

16 A contrast nicely challenged by Sedley, -Epicurean Anti-Reductionism," in J. Barnes and M. Mignucci, edd., Matter and Metaphysics (Naples 1989) 297-327.

17 M. Nussbaum gives a particularly sinister reading of Epicurean moral education, based primarily on Philodemus' reports: -Therapeutic Arguments: Epicurus and Aristotle," in M. Schofield and G. Striker, edd., The Norms of Nature (Cambridge 1986) 31-74. I believe that she both underestimates the rationality of Epicurean practice and overstates the focus on reason in Aristotelian moral formation to heighten her contrast; but the question is beyond the scope of this essay.


DA VID K. O'CONNOR 173

also remind us of our belief and give us their support and example. We live the Epicurean way of life most fully and with least risk of straying from the path when we have the encouragement and scrutiny of our fellow travelers.

II. Does Epicurean Friendship Undermine Self-sufficiency?

I have reconstructed the Epicurean ideal of friendship as a kind of safe harbor from the worldly storms-physical and mental-that threaten to disturb human tranquillity. But the fashion recently has been to suspect that this harbor is mined with disturbing elements of its own. In particular, a number of recent commentators have emphasized the ways in which valuing close personal attachments can make our happiness vulnerable to various kinds of chance disruption.1s This seems to create difficulties for Epicurus' moral theory, for he clearly places a high value upon both friendship and self-sufficiency. Since self-sufficiency requires that one's happiness be relatively (some would say absolutely) resistant to chance disruption,19 is it not impossible to combine in a single coherent moral ideal the demands for both friendship and self-sufficiency?

I believe the Epicurean would have good cause for rejecting this inference. There is no general reason to expect friendship to undermine self-sufficiency; everything depends on what kinds of friendship we are talking about, and on exactly how chance is supposed to threaten them. I shall consider two ways in which friendship might be thought to expose us to chance disturbances, and try to show that the Epicurean ideal is not threatened by either of them. First, the very formation of friendships might be a matter of chance. Perhaps they spring up spontaneously when people of complementary temperament come together, independent of any calculated attempt to seek them out. 20 If friendship is both spontaneous and indispensable for a happy life, rational control over happiness will be limited. We may not happen to meet enough of the right sort of people to satisfy our need for friendship. Second, the chance of the

18 E.g. Mitsis, Nussbaum, and Williams (supra nn.1-2).

19 By no means do I think this captures everything, or even the most important thing, that Greek moralists had in mind when they valued self-sufficiency (o.u't(XpKEto.). But the issue of vulnerability to chance is important, and it has been prominent in recent discussions.

20 Something resembling this view is present even in Cicero Fin. 2.83 on the contrast between calculated and spontaneous friendship.



dissolution of a friendship might be thought to expose us to pain. Sometimes these dissolutions may result from competition or betrayal among friends; but the most final dissolution of all occurs at death. For humans who have no control over death, does not friendship necessarily expose our happiness to the pain of separation and mourning?

In these two ways, chance might be thought to threaten friendship at its beginning and end. How can Epicurus preserve us from its risks without denying us its blessings? Let us first consider the alleged spontaneity of friendship's beginnings. The picture of friendship as a delicate plant that springs up spontaneously only under quite special conditions of mutual compatibility may be appropriate for certain kinds of personal relationships among contemporary bourgeois academics,21 but it seems much less applicable to the sort of supportive fellowship at the heart of Epicurus' ideal. Though they were certainly warm and intimate, Epicurean friendships were not, I suspect, 'personal' in the way some of our friendships are. Cicero's Epicurean spokesman Torquatus (Fin. 1.65) contrasts the large number of friends who lived with Epicurus with the exclusivity of the famous, intensely personal relationships (e.g. between Orestes and Pylades) usually held up as paradigms of friendship. Whatever spontaneous combustion may have been required to weld together these legendary pairs, there is little reason to think that the broader bonds of fellowship have such volatile and unpredictable beginnings.

The bonds of an Epicurean community are more characteristic of friendly fellowship than of the personal intimacy presumed by the 'spontaneity' view of the formation of relationships. "We should laugh, philosophize, and handle our household affairs and other personal matters, all at the same time," says Epicurus (VS 41, tr. Long/Sedley). Such a community is not a product of chance: we build it up on our own initiative. There is no threat to self-sufficiency to be found here.

But even those who accept my suggestion that the beginning of Epicurean fellowship is not much exposed to chance disturbance may think that its end cannot avoid being so. In the first place, the everyday threats to friendship of betrayal and competition can

21 I am actually skeptical that the description fits even these personal friendships. If the conditions are so much in the hands of fortune, why are they so often fulfilled?


DA VrD K. O'CONNOR 175

shatter an Epicurean fellowship as much as any personal friendship. Being part of a fellowship, on this view, makes us vulnerable, and contradicts the demand of self-sufficiency for invulnerability. The surviving evidence indicates that Epicurus was indeed somewhat wary of friendship, but he offered advice that would minimize the risks and make friendship among Epicureans secure.22 But his most important way of making people into good and secure friends was by making them into good and secure Epicureans, with Epicurean tastes and goals. By rooting out the greed (VS 43) and political ambition (e.g. KD 7) that are the underlying cause of the greatest discord among friends,23 Epicurus makes his followers the kind of people who are not risky friends. The notion that friendship necessarily makes us vulnerable to betrayal or other hurt by our friends 24 seems much more at home in a modern personal friendship than in the moderate fellowship of Epicureanism.

But death is another story (VS 31): "Against all else protection can be provided, but because of death all human beings dwell in an unfortified city." Not even Epicurus could find a way to shield friendship from its unpredictable and inevitable disruptions. Our friends not only may die, they will die. Must we drink this wormwood wi th the honey?

Epicurus did not think so, and tried to find a response to the death of friends that would be unmarred by emotion and irrationality. His suggestion is summarized by this saying (VS 66): "We should feel for (auJl1ta900JlEv) our friends not through mournfulness but through mindfulness (cpPov"ci~OV'tEC;)." We do not know as much as we would like about the concrete embodiment of this attitude in Epicurean community, but I believe we should look to the way Epicureans remembered the exemplary lives of the dead in the cultic celebrations. Epicurus must have felt very strongly the contrast between the tears and wailing of traditional mourning and the calming and pleasant remembrance of past intimacy that he recommended.

What made this radically new and undisturbed acceptance of the passing of friends possible was, of course, the famous claim that

22 For advice, see VS 39; for Epicurus' wariness, along with a final judgment in favor of friendship's prudence when proper safeguards are respected, VS 28.

23 Aristotle (esp. Eth.N ic. 9.8) rather than Epicurus is the explicit source of this claim about the causes of discord.

24 See e.g. Mitsis 150 with n.46.



"Death is nothing to us." This was the final and greatest protection against the slings and arrows of fortune that an Epicurean could have, and explains how Epicurus' friend Metrodorus could say (VS 47), "When it is time for us to go, we spit on life." Two features of the Epicureans' attitude to their own deaths are especially important for removing the sting from the death of friends as well. First, after death there is no sensation, so no disturbing pains; so there is no reason to worry about one's state after death (KD 2). Second, "an unlimited time and a limited time have equal pleasure"; so there is no reason to regret the deprivation of future pleasures at the cessation of life.25 Epicureans have no more reason for worry about the pains of their departed friends or for regret at their friends' deprivation of future pleasures than they have for worry and regret about their own deaths. They might still regret being deprived of the pleasure of a friend now dead; but Epicurus' emrlhasis on the pleasures of memory, and the practical reanimation 0 such memories in the Epicurean cult, seem designed to remove even this source of disturbance (VS 75). The joys of a departed friend need not be irrevocably past. Once again, I think we will better appreciate this security of friendship even in the face of death if we keep in mind the Epicurean ideal's focus on fellowship rather than our more personal style of friendship.

To sum up, I do not think the friendly fellowship of the Epicureans made them vulnerable to chance disturbances in a way that is inconsistent with their concern for self-sufficiency. Fellowship does not depend to any important extent on rare, spontaneous attachment; it is not particularly exposed to dissolution by betrayal and competition, especially among people with the tamed desires for wealth and honor characteristic of good Epicureans; and Epicurus took care to make the pleasures of friendship secure even against death. The notion that friendship must make us vulnerable to such chance threats to our happiness assumes, if only implicitly,

25 KD 19. I believe these two doctrines are the -judgment" referred to in KD 28 that guarantees the security of friendship from eternal or even long-lasting pain; but the saying is rather obscure. The nature and justification of the Epicurean attitude to death (on which see most recently P. Mitsis, -Epicurus on Death and the Duration of Life," Proceedings of the Boston Area Colloquium in Ancient Philosophy 4 [1988] 295-314) is of course a complicated and controversial topic that I cannot address here. I aim only to clarify how death does not make friendship and self-sufficiency incompatible for an Epicurean.



a model more appropriate to contemporary personal friendship than to the supportive fellowship recommended by Epicurus.

III. The Ancient Criticism: Epicurean Friendship is Ignoble

So far I have concentrated on showing that Epicurus' ideal of friendship is consistent with his hedonism and his concern for selfsufficiency. In doing so, I have remained largely inside the Epicurean perspective, taking for granted the supreme importance of pleasant tranquillity and asking merely how Epicurean friendship will help us achieve it. In the rest of this paper, I want to consider two criticisms of the Epicurean ideal that originate outside this basic hedonist perspective. First, I want to examine the charge that the Epicurean ideal cannot justify dying for a friend; in the following section, I consider the charge that Epicurean friendship is overly egoistic. Despite appearances, these two objections have very different grounds, and we must understand the ideals implicit in them before we can see how Epicurus might respond. I am at least as concerned to clarify what is at stake in these criticisms as to decide who wins the argument.

Through much of the second book of De finibus, Cicero focuses his critique on the Epicurean denial that virtue and nobility are rationally attractive independent of pleasure. Cicero sees this denial as the central weakness of Epicureanism, and believes that if he "can succeed in proving the existence of the noble (honestum) as a thing essentially and for itself desirable" (Fin. 2.44, tr. Rackham [modified]), he has refuted the entire Epicurean teaching. Within the scope of the noble Cicero seems to include both intellectual and moral excellence, though by far his greater emphasis is on the latter. 2l, More specifically, Cicero (esp. 2.45-47) sets out to show that reasonable people will find the four cardinal virtues attractive and even beautiful in their own right, whether or not they bring any advantage or reward. He argues this point in detail for justice, courage, and temperance. A recurring feature of these arguments is the presentation for Torquatus' consideration of famous examples of virtuous behavior under unpleasant, disadvantageous, and dangerous conditions. Trusting more to Torquatus' solid Roman character

26 For the inclusion of intellectual excellence in the noble, see especially his reference at Fin. 2.111 to the pleasure-independent attraction of honestissima studia.



than his Epicurean allegiance,27 Cicero is sure that his interlocutor will feel the attraction of virtue even in these cases, proof that the attraction is independent of pleasure.

Prominent among Cicero's examples of the independent attraction of the noble are cases in which certain Romans gave up their very lives for the sake of virtue. These cases are not intended primarily to illustrate the altruism of the actors. 28 Cicero's entire emphasis is on how such people must have aimed at something other than mere pleasure, something Cicero argues is of higher rank and independently desirable. The courage of soldiers who boldly go to their deaths in battle is perhaps the most typical example of a virtue that he claims to be inexplicable in terms of pleasure. But Cicero gives examples of the other virtues as well. Marcus Regulus of his own free will departed from Rome, where he could have lived a famous and respected man, to return to captivity in Carthage because of a pledge he had given to an enemy. His concern for his own integrity and fidelity (aspects of justice in Cicero's view: 2.46) easily outweighed for him all the attractions of a peaceful and honored retirement. Lucretia committed suicide to defend her temperance, outraged by sexual mistreatment by a tyrannical ruler. In all these cases, acceptance of death shows a depth of commitment to a sometimes painful nobility that, Cicero believes, belies the Epicurean reduction of the rationally desirable to the pleasant.

I propose that Cicero continues this general line of attack when he turns to criticize the Epicurean account of friendship. This critique immediately follows the section on the independent attractiveness of the virtues, and a number of passages show the close links between the two discussions. 29 He is simply applying to the

27 Cicero (2.58) emphasizes that both Torquatus' and Epicurus' professed doctrines neither justify nor do justice to the nobility of their lives.

28 Of course some of Cicero's exemplary noble actions did in fact benefit people other than the actor. Courageous military deaths benefit fellow soldiers and citizens (2.60-62); Marcus Regulus' justice respects a pledge to the enemy (65); Lucretia's suicide in defense of her sexual temperance was instrumental in procuring Roman liberty (66). But Cicero does not emphasize that such actions benefit others more than the actor so much as that they manifest more regard for the noble than for pleasure. He contrasts acting for the sake of the noble with acting for one's own pleasure, not with acting for one's own pleasure.

29 See especially 2.83. Cicero is criticizing the view of some later Epicureans that friendship involves a contract to feel toward a friend as one feels toward oneself: • If men have succeeded in making this contract, let them make a further contract



nobility of friendship the same sort of considerations that he had earlier applied to the cardinal virtues. Once again, he presents to Torquatus an example of noble action in a situation where pleasure cannot be the motivation, and trusts to Torquatus' good character to draw the conclusion that nobility can be choiceworthy independent of pleasure. And as with the virtues of courage, justice, and temperance, this independence is illustrated best when an actor faces a noble death for friendship's sake.30 Cicero (2.79) gives two such illustrations: the devotion of Phintias to Damon and of Pylades to Orestes.

This 'Pyladic' type of friendship (2.84) contrasts with friendships that aim at profit or advantage. Cicero (2.85) describes this contrast by saying, "You must love me for myself, not my possessions, if we are to be genuine friends." We might take this as primarily acontrast between egoistic and altruistic friendship, but I think the context, especially the parallel with what Cicero says about the virtues, indicates that this is not what he had in mind. He is trying to show that Epicurean friendship, like Epicurean virtue, is ignoble because it aims at low ends like profit or pleasure rather than at nobility itself. 31 Devotion to a friend 'for himself' even when no pleasure results shows that one is motivated by the independent attraction of nobility, just as the deaths of Regulus and Lucretia showed their noble devotion to justice and temperance. No doubt there is something selfish about Epicurean friendship as Cicero understood it; but in keeping with his general criticism of Epicureanism, he was more exercised by its ignobility than its egoism.

Cicero's argument for the ignobility of Epicurean virtue and friendship silences the Roman Epicurean Torquatus, but all the important moves in the argument appeal more to Romanness than

to love fair-dealing, self-control, and all the virtues for their own sakes and without reward" (tr. Rackham [modified]). Epicurean hedonism threatens the independent value of friendship in the same way it threatened the independent value of the virtues. See also 2.81, where Cicero claims that the actual concern for the noble in Epicurean friendship is inconsistent with the professed value of pleasure, and 2.85, where in the conclusion of the friendship section he links it to the virtue section.

30 Cf Arist. Eth.Nic. 1169a18-22 on why a friend of the best kind would be willing to die for a friend: such a man .. acquires for himself the noble {'to xaAOv)."

31 This interpretation is confirmed by Cicero's similar critique of hedonic friendship in Arnie. 9.32 and 15.52.



Eficureanism. Torquatus does not dispute any of Cicero's examples o noble action; he shares the acceptance of patriotism and political ambition that lies behind Cicero's claims. Obviously this is a bizarre thing for an Epicurean to do. We can construct the outlines of a more authentically Epicurean response fairly easily. Epicurus was perfectly clear in his rejection of the pleasure-independent desirability of the very sorts of activities that Cicero presents as paradigms of virtue. He despised political ambition, and warned his followers (e.g. KD 7, 14; VS 58) to release themselves from "the prison of politics." He also cautioned his followers (e.g. KD 11, VS 45) not to be inflated by the pretensions of philosophy. Epicurus, then, offered a general critique of the independent attractiveness and nobility of just the sorts of intellectual and moral virtue to which Cicero so freely appeals. In the light of this critique, it is an important and desperate puzzle how a public-spirited Roman like Torquatus ever thought of himself as an Epicurean at all.32 But for Epicurus himself there is no puzzle; Cicero's arguments can get no grip on him.

Since Epicurus would reject the assumption of the pleasure-independent rational desirability of the noble on which Cicero builds his critique, he need not be disturbed by failing Cicero's 'nobility test' for friendship. Epicurean fellowhip will not be 'Pyladic', just as Epicurean virtue will not find in the ambitious courage of a soldier, the absolute devotion to justice of a Regulus, or even the outraged honor of a Lucretia, much of a model. Of course this does not mean that a consistent Epicurean would never die for a friend; some such deaths may be justified on hedonic grounds. Epicurus himself nearly lost his life in a shipwreck when he ran the risk of visiting his friends in Lampsacus.33 But this is a far cry from the

32 Diskin Clay has some very interesting suggestions about how Lucretius adapts the Epicurean message to a Roman audience: Lucretius and Epicurus (Ithaca 1983). The general issue of what practical meaning allegiance to Epicureanism had for political Romans deserves serious consideration. It is curious that Cicero argues for the anti-political implications of Epicurean hedonism as if these implications were not explicitly acknowledged by the Epicureans themselves. Why does he not simply confront the public-spirited Torquatus with the explicitly anti-political statements of the Master?

33 See D. Clay, ·Sailing to Lampsacus: Diogenes of Oenoanda, New Fragment 7," GRBS 14 (1973) 49-59, for the evidence of this shipwreck and Epicurus' use of it in his moral teaching. Could this be one of the risks Epicurus had in mind when he said (KD 28) that we must run some risks for friendship'S sake? Clay points out


DA VID K. O'CONNOR 181

heroic deaths faced by legendary friends like Pylades or Achilles. The Epicurean indifference to fame and death makes a poor subject for epic poetry. As Cicero (Fin. 2.97, tr. Rackham) says, "A great commander's death is famous, but philosophers [like Epicurus] mostly die in their beds. »

These considerations open up onto a broader and more speculative set of issues about what we might call the virility of the Epicurean ideal of friendship. To a Roman like Cicero, there was something unmanly about Epicureanism, and its rejection of political friendship was, I think, a particularly galling feature of this effeminacy.34 How could a real man be satisfied with such domesticated pleasures? Where is his sense of adventure? To a certain extent, Epicurus' critique of nobility is also a critique of virility and its demands. The suspicion generated by the Epicureans' indiscriminate mixing across political, social, and sexual boundaries may reflect an uneasiness at their blurring of sexual identity, a disgust at a kind of moral transvestism. The Epicurean way of friendship made men too soft and women too public in a society where hardness and privacy were defining characteristics, respectively, of the male and the female.35

In the end, the challenge 'Will an Epicurean die for a friend?' cannot be answered straightforwardly. There are probably cases in which an Epicurean would indeed choose to die for a friend; but this choice looks much more significant to someone with Cicero's starting points than Epicurus'. Cicero thinks such extreme cases show how nobility shines through with its own value, independent of pleasure; Epicurus would find them no more than curious illustrations of his general teaching on pleasure and death. They are not very revealing test cases of anything. Epicureanism rejects the very understanding of nobility that gives Cicero's question its point.

how Epicurus models his account of the shipwreck on a passage of the Odyssey; it is hard to imagine any Epicurean heroism that could be modeled on the Iliad.

34 E.g. Arnie. 13.47, where Cicero contrasts the risks taken in manly friendship with the craving for security of friendships fit only for women, the poor, and the unlucky. For criticisms of the virility of Epicureanism, see Fin. 2.47, 73, 94f. Aristotle, e.g. Eth.Nic. 1 09sb 19-22, also attacks hedonism as fit only for an unmanly, servile person (avopa1toorooTl~).

35 Paul Veyne's discussion in The History of Private Life I (Cambridge [Mass.] ~ 987) of sexual identity in the ancient world has stimulated my thinking on these Issues.



IV. The Modern Criticism: Epicurean Friendship is Egoistic

On my reading, the Epicurean ideal of friendship emphasizes fellowship rather than altruism. This conclusion will trouble many readers. Does Epicurean friendship require, or even allow, the altruistic concern for another for that other's sake that is at the heart of real love and friendship? Does not Epicurus' hedonism, granting ultimate value only to pleasant psychological states of individuals, commit him to an unacceptably selfish account of friendship? I answer that in the Epicurean ideal of fellowship reconstructed above (section I) there is no obvious element of selfishness; and the mutual support and enjoyment of such friends seem attractive and generous. This generous sociality is grounded in an Epicurean education of desire, especially in a critique of greed and ambition, rather than in any direct cultivation of altruistic concern.

Some will not be satisfied by this, and will insist that friendship without altruism is inherently defective. They will either have to give up on Epicurus, or try to find room within his generally egoistic hedonism for altruistic friendship. Mitsis has recently taken this second path, arguing that there are ineliminable altruistic commitments in the Epicurean account. 36 He believes that, like John Stuart Mill, Epicurus was forced to admit into his hedonist theory "non-hedonistic grounds of preference,» but was unaware that he had done so because he had accepted "an illegitimate move from 'pleasurable pursuits' to 'pleasures.'»37 I do not think Epicurus made this illegitimate move,38 but for present purposes I want to concentrate on the evidence for ascribing to him an altruistic view of friendship.

36 See Mitsis 129f with n.7. 37 Mitsis 132, 141, both times referring to Sidgwick's criticisms of Mill in Meth

ods of Ethics.

38 Besides his claim that Epicurus allows non-hedonistic grounds of preference by valuing altruistic friendship, Mitsis (142 n.31) mentions only one other aspect of his moral theory that this alleged illegitimate move infects: "'Epicurus relies on a non-hedonic criterion when defending his [preference for] pleasures which are completely up to us (1tap' ftjl.a~) and not subject to chance. JJ I see no reason to accept this. Self-sufficiency surely can be defended on hedonic grounds; it minimizes the likelihood of chance disturbance and increases our enjoyment of luxuries when we do happen upon them: see Ep. ad Men. 13of. If Epicurus nowhere else makes the mistake about pleasure that Mitsis alleges he makes regarding friendship, the allegation is not very well supported.



Mitsis' argument hinges on two passages. The first is from De finibus, consisting of the Epicurean Torquatus' first of three arguments against the charge that hedonism destroys the possiblility of friendship.39 Two questions must be answered about this passage: (1) Does it ascribe altruism to friendship? and (2) Is it good evidence for Epicurus' view rather than (like the other two arguments) merely for the views of later Epicureans?

I think the answer is <No' to both, though the structure of the argument is so obscure that it is hard to be confident about the first question. The argument tries to show the consistency of friendship with the position that "the pleasures which belong to friends are not as desirable per se as those we desire as our own" (Fin. 1.66, tr. Long/Sedley). I agree with Mitsis 40 that this position accurately reflects "Epicurus' widespread commitments to egoism and hedonism: and take it as an accurate statement of Epicurus' own view. Critics like Cicero (1.66, 69; 2.80) claimed that this view would make friendship "unstable" or "completely crippled: and apparently some later Epicureans (whom Torquatus calls "timid") agreed. They in effect gave up the position to accommodate the critics' demand that friends "be loved for their own sake even if no advantage accrues from the friendship. "41 Cicero reports their views in Torquatus' second and third arguments. But other and presumably bolder Epicureans tried to accommodate the critics' demand without giving up the authentic view of· Epicurus, and Torquatus' first argument was the result.

The argument runs as follows: (1) Premise: friendship is necessary both to create and to protect the pleasant security of a happy life. (2) Premise: we cannot preserve friendship "unless we love our friends as much as ourselves" (1.69, tr. Long/Sedley). (3) Conclusion: "the wise man will have just the same feelings towards his

39 Fin. 1.65 (cf 2.78) states the charge, while 1.66-70 provides Torquatus' three arguments. Mitsis takes only the first argument (66-68) to be relevant to establishing that Epicurus (as opposed to later Epicureans) held an altruistic view of friendship, since Cicero's reply to the second and third arguments (2.82) explicitly denies that they are from Epicurus.

40 Mitsis 139f n.26. J. M. Rist, Epicurus: An Introduction (Cambridge 1972) 129, claims that -Torquatus does not attribute this view to Epicurus himself and the context makes it certain that it was not the master's position," but I see no reason to accept this.

41 1.69 (tr. Long/Sedley). Cicero (esp. 2.82) makes this clear in his reply to the second and third arguments.



friend that he has for himself, and he will work as much for his friends' pleasure as he would for his own'" (1.67, tr. Long/Sedley). This conclusion is supposed to show that the wise Epicurean will fulfill the critic's demand about friendship while remaining true to the authentic view of Epicurus that our own pleasures are in themselves more desirable than those of friends. Unfortunately there is a glaring weakness in the argument, centering on the crucial premise (2). The distinction between finding our friends' pleasures as desirable in themselves as our own (which is ruled out by Epicurus' authentic view) and loving our friends as much as ourselves (as required by the argument's conclusion) is left utterly obscure. If they collapse into the same thing, Epicurus' position is abandoned as in the second and third arguments. But if the distinction is preserved, it is not clear whether or not the resulting position will require altruism.42 I confess that I find the argument incoherent, and doubt that any conclusions can be drawn from it about how the bolder E1?i~ureans remained true to Epicurus while accommodating the cntlcs.

There is, I think, is good reason to think that this lame argument is no reflection on Epicurus. Torquatus says that all three arguments come from "us,'" i.e., from us later Epicureans. This contrasts with the way he introduces his account of friendship by referring to a statement of Epicurus: "Epicurus' pronouncement about friendship is that of all the means to happiness that wisdom has devised, none is greater, none more fruitful, none more delightful than this. "'43 Similarly, Cicero (2.82, tr. Rackham) in his response refers to "a saying of Epicurus himself-that friendship cannot be divorced from pleasure, and that it deserves to be cultivated for the reason that without it we cannot live secure and free from alarm, and therefore cannot live agreeably.'" I think authentic sayings of Epicurus like these, which emphasize the hedonistic justification of friendship, are being interpreted by "us (later) Epicureans'" in Torquatus' first argument, just as in the other two. Thus even if this

42 I do not claim that the very idea of loving a friend as much as oneself is incoherent; but unless we know more precisely what T orquatus' bolder Epicureans had in mind by this, we do not know whether such loving would really be altruistic. The bare statement of the second premise gives us no clue.

43 Fin. 1.65 (tr. Rackham). Torquatus also summarizes the point of the first argument with a translation of KD 28, which I take as further proof that the argument itself is intended to interpret Epicurus, but is not based on his explicit statement.



argument could be shown clearly to espouse altruism (which I doubt), we have no reason to think it represents Epicurus' own view rather than a later modification introduced under the pressure of criticism.

The second passage to which Mitsis appeals is indeed an authentic statement of Epicurus, albeit an emended one: "All friendship is choiceworthy for itself (bt' rau'tTtv); it has its beginning (apXTt) from utility (a1tO 'tile; rocp£AElae;). "44 Does this saying "strongly suggest an unselfish picture of friendship" and "enjoin [aJ positive, altruistic concern for others' interests"?45 Notice first that it does not say that a friend is choiceworthy for himself, whatever that may mean; it says that friendship is. Mitsis sometimes seems to conflate these two notions, as if attributing a value to friendship independent of pleasure would be equivalent to attributing a value to my friend independent of my egoistic interests. Of course it would be inconsistent for Epicurus to hold both that pleasure is the sole thing rationally desirable and that friendship is rationally desirable independent of its pleasure; but even this would not commit him to altruism. Consider an illustration of the difference: a political man might claim that kingship is choiceworthy in itself without reference to its pleasure or utility, but deny that the king's subjects should be valued altruistically at all.

If, then, this passage is evidence of an inconsistency in Epicurus, it concerns the reduction of the rationally desirable to the pleasant, not egoism and altruism. As I argued in the previous section, Cicero was primarily interested in the first issue; but Mitsis is primarily interested in the second. But I do not think we should interpret this isolated saying in a way that so blatantly contradicts Epicurus' hedonism unless there is no reasonable alternative. One possibility is Rist's suggestion that Epicurus does not mean that friendship is valuable independent of its pleasure, but only that it is directly pleasant, without intermediaries. 46 Let me suggest another alternative reading. This saying is usually translated with a strong contrast

44 VS 23 (quoted supra n.8). With Mitsis I accept Usener's change of apuTt to ai.pE'tTt·

45 Mitsis 129f. 46 Rist (supra nAO) 132. Rist draws a parallel with the Stoic conception of 'things

preferred' (npol1'YJ.LEva), some of which, while not valued independently of virtue, were said to be preferred for their own sake.



between "in itself" and "from utility. "~7 But there is no textual necessity for this. Equally, we may take the second clause to explain or amplify the first rather than to contrast with it. Epicurus would then be saying roughly that all friendships are in themselves good things to have because they have been founded in utility. Perhaps he had in mind a contrast between the general case of friendship being choiceworthy in itself, and exceptional cases in which it should nonetheless be avoided. He considers a parallel issue with regard to pleasures in general (Ep. ad Men. 129f; KD 8): though a pleasure is in its own nature something good, there can be exceptional occasions when we should refrain from choosing it. This gives a natural reading to the saying without attributing to Epicurus a stark contradiction between his hedonism and VS 23.

I do not think that Mitsis' attempt to save Epicurus from the charge of egoism is successful. But I want to conclude by suggesting that Epicurus' egoism is in this regard unobjectionable. For even if Epicurus' ideal of friendship is egoistic, it is not selfish. Once again it helps to consider what concrete life as an Epicurean would be. There will be no exploitative relationships bent on extracting money or power from others, for Epicurus has taught us that greed and ambition are empty and vain-they will distract us from our real goal of peaceful tranquillity. There will be the fellowship of a community that remembers its departed saints and thereby supports its living members. And there will be communal meditation on the philosophy that justifies this way of life. What more could altrusim demand of us than this? The solidarity of our fellowship eclipses the altruism of the individual. "Friendship dances round our dwelling, inviting us all to awaken to the call of blessed happiness. "~8

UNIVERSITY OF NOTRE DAME August. 1989

47 E.g. Long and Sedley connect the two clauses with -but." 48 VS 52. I would like to thank the other participants in the conference for their

fellowship and discussion, especially Phillip Mitsis and Paul Vander Waerdt for their extensive written comments on an earlier draft.

The Invulnerable Pleasures of Epicurean Friendship

Documents

Transcript of The Invulnerable Pleasures of Epicurean Friendship